I found and reported an authentication failure in the WPForms PayPal Commerce webhook, the webhook route being public was not the vulnerability as webhooks have to be publicly reachable so that PayPal can deliver events.
The problem was what happened after the request arrived. In affected versions, the handler could process a supported event before establishing that PayPal was actually the sender. In my local lab, a forged event could change the state of a matching payment record.
The expected order is:
The affected flow effectively performed steps 2 and 3 without first completing step 1. The issue was fixed in WPForms 1.10.0.5 and is tracked as CVE-2026-4986.
Then came the part I found more interesting: triage told me I was reporter #11. That number does not prove exploitation, and it does not tell us the total number of people who found the vulnerability. It does establish a lower bound: at least eleven researchers independently converged on the same trust failure.
The write up covers:
- the vulnerable code path
- my local reproduction
- why payload validation was not sender authentication
- the fallback listener
- the patch
- why duplicate reports may be useful rediscovery intelligence
Full write-up: https://blog.himanshuanand.com/2026/07/reporter-11-10-people-found-the-wpforms-paypal-bug-before-me-cve-2026-4986/
Testing was limited to my own local environment. I am not claiming original CVE credit; I independently rediscovered and reported the issue. Disclosure: I wrote and performed the research, code review and local reproduction.
I used an AI to help copy edit and organize the final article.
Should duplicate report volume affect how urgently a vendor treats a vulnerability?
Login